Fault detection in electromechanical systems plays a significant role in product quality and manufacturing efficiency during the transition to smart manufacturing. Because collecting a sufficient number of datasets under faulty conditions of the system is challenging in practical industrial sites, unsupervised fault detection methods are mainly used. Although fault datasets accumulate during machine operation, it is not straightforward to utilize the information it contains for fault detection after the deep learning model has been trained in an unsupervised manner. However, the information in fault datasets is expected to significantly contribute to fault detection. In this regard, this study aims to validate the effectiveness of the transition from unsupervised to supervised learning as fault datasets gradually accumulate through continuous machine operation. We also focus on experimentally analyzing how changes in the learning paradigm of the deep learning model and the output representation affect fault detection performance. The results demonstrate that, with a small number of fault datasets, a supervised model with continuous outputs as a regression problem showed better fault detection performance than the original model with one-hot encoded outputs (as a classification problem).

본 연구는 다시마 양식을 위한 통합 자동화 시스템을 개발하고, 이를 통해 생산성, 비용 효율성, 환경적 지속 가능성을 모두 개선하는 데 중점을 두고 있다. 기존의 노동 집약적 수확 방식과 넓은 공간을 필요로 하는 수평 건조 방식은 비효율적이며, 환경적 부작 용을 초래했다. 이에 본 연구는 자동화된 수확 시스템, 해상-육상 연계 운송 시스템, 그리고 수직 건조 시스템을 통합적으로 개발하여 양 식업의 생산성을 극대화하고 자원 사용을 최적화하였다. 자동화된 수확 시스템은 작업 속도를 약 35% 향상시켰으며, 작업의 일관성을 유지하여 품질 오차율을 2% 이하로 줄이는 성과를 보였다. 해상-육상 연계 운송 시스템은 모듈형 컨테이너를 활용하여 운송 중 손상률 을 기존 15%에서 5%로 감소시켰고, 운송 시간을 평균 6시간에서 4시간으로 단축하였다. 또한, 수직 건조시스템은 고밀도 적재와 자연 대류 방식을 도입하여 건조 시간을 기존 48시간에서 28시간으로 40% 단축하였으며, 에너지 소비를 25% 감소시켰다. 이러한 시스템은 데이터 기반으로 설계 및 검증되었으며, 통합적으로 양식업의 경제성 향상과 환경적 부담 감소를 동시에 실현하였다. 본 연구의 결과는 다른 해조류 양식에도 적용 가능하며 지속 가능한 해양 자원 관리에 기여할 것으로 기대된다.

The demand for automated diagnostic facilities has increased due to the rise in high-risk infectious diseases. However, small and medium-sized centers struggle to implement full automation because of limited resources. An integrated molecular diagnostics automation system addresses this issue by integrating small-scale automated facilities for each diagnostic process. Nonetheless, determining the optimal number of facilities and human resources remains challenging. This study proposes a methodology combining discrete event simulation and a genetic algorithm to optimize job-shop facility layout in the integrated molecular diagnostics automation system. A discrete event simulation model incorporates the number of facilities, processing times, and batch sizes for each step of the molecular diagnostics process. Genetic algorithm operations, such as tournament, crossover, and mutation, are applied to derive the optimal strategy for facility layout. The proposed methodology derives optimal facility layouts for various scenarios, minimizing costs while achieving the target production volume. This methodology can serve as a decision support tool when introducing job-shop production in the integrated molecular diagnostics automation system

Fishways, particularly installed at the estuary, have a purpose to encourage fluent migration for migratory fishes, as well as amphidromous and even freshwater species. Not choosing the laborious traditional method of using traps, we assessed the efficiency of the two fishways installed at the west and east barrage of the Nakdong River estuarine barrage respectively, by analyzing the videos recorded through automatic monitoring system. We randomly selected 30 videos monthly at each fishways and identified what kind of fishes were using the fishways and categorised their behaviour such as size, time and whether they passed the monitoring system or not. As a result, a total 8 families 14 species were recorded by monitoring system, with the most dominance of Erythroculter erythropterus (Relative Abundance: 59.5%), followed by Micropterus salmoides (R.A: 19.9%) and Mugil cephalus (R.A: 9.9%). The monitoring system can capture passing fishes during night but the number of appearances of fish species at each hour of a day indicated significant diurnal activities (p<0.05). When fishes pass the monitoring system, approximately 70% of them passed through the monitoring device, while 17% of them showed fallback movement. Our finding indicates that species-specific characteristics of each fish are well represented through video monitoring method. In order to maximise advantages of using video monitoring, it is necessary to consider the installation point properly so that the monitoring system does not interfere with the movement of fish. Also, the utilisation of AI technology in the future is also necessary.

In this study, fire extinguisher system to which form fire extinguisher agents were adopted was applied to the combat vehicle crew room to apply fire extinguishing performance and acid gas safety that meet the national defense standards. As a result of evaluation and verification, the following conclusions were drawn. For standard fire sizes in the combat vehicle crew's standard model, we ignited using a mixture of Novec 1230 and Halon 1301 form extinguisher agent and released form extinguisher agent after 30 seconds to determine the fire extinguishing time. The amount of acid gas generated met the criteria in all cases. When the fire size was increased to 0.12m2 and a 2.0mm nozzle was used, all of the extinguishing time, the amount of acid gas generated, and the concentration of Novec 1230 met the criteria. Despite the more difficult conditions to extinguish the fire by making the fire larger, it was possible to confirm the extinguishing performance of the Novec 1230 form extinguisher agent and its safety against acid gas.

To efficiently develop an automatic assembly system that can enhance the quality and assembly productivity of the shaft assembly in EV relays, a DMU model was utilized. After modeling each component of the assembly system using the CAD software CATIA, a DMU model of the assembly cells and the entire assembly system was created using the assembly model. Additionally, the DMU Kinematics Workbench was employed to verify and validate the design of the automatic assembly system for the shaft assembly, a key component of the EV relay, before actual construction. This approach helped reduce time and costs by minimizing trial and error.

교통안전시설물의 관리는 도로교통의 안전과 직결되는 문제이다. 운전자는 신호등, 표지판, 노면표시 등을 통해 운전에 필요한 정보 를 얻는다. 노후된 표지판과 노면표시는 운전자에게 잘못된 정보를 제공할 수 있으므로 주기적인 시설물의 관리가 필요하다. 본 연구 는 딥 러닝 기술을 활용해 운전자 시각의 영상 자료에서 교통안전표지를 자동으로 탐지하고자 하며, 교통안전표지의 공통된 색상 특 징을 기반으로 클래스를 그룹으로 묶어 데이터셋을 구축하는 방법을 제안한다. 객체탐지의 성능지표로 널리 활용되는 mAP를 사용해 클래스 묶음 여부에 따른 탐지 성능을 비교한 결과, 색상 기반 클래스 묶음을 적용한 모델의 탐지 성능이 비교군에 비해 약 36% 상승 함을 확인하였다.

현재 존재하는 인공지능 기반 음악 생성에 관한 여러 모델과 연구는 수동 텍스트(Text) 기반 음악 생성에 대해 다루고 있다. 본 논문은 사용자의 편의성을 높이고, 창의적인 음악 생성 과정 을 더욱 원활하게 할 수 있도록 텍스트(TEXT) 프롬프트(Prompt) 자동화를 통한 음악 생성시 스템 방안을 제안한다. 그 방안으로 음원 파일을 통해 수집한 음악 분석 및 데이터화와 가사 정보에서 추출한 키워드를 기반한 장르, 가수, 앨범 등의 정보가 포함된 데이터셋(Dataset)을 구축 후, 파이썬(Python)의 자연어 처리 방법인 Konlpy를 사용하여 가사 데이터를 토큰화하고, TF-IDF(Term Frequency-Inverse Document Frequency) 벡터화를 통해 중요한 단어를 추 출한다. 또한, MFCC, 템포 등의 특징 데이터셋을 통하여 모델을 통한 감정을 예측하고, CNN 모델 및 Chatgpt를 활용한 텍스트 프롬프트를 자동생성하는 방법을 구현하여, MusicGen 모델 을 사용한 자동화 생성 프롬프트 기반 음악을 생성한다. 본 텍스트 프롬프트 자동 생성 화를 통한 음악 생성 연구의 결과는 음악 데이터 분석 및 생성 분야에 기여될 것으로 기대한다.

In this study, we investigated the time signal devices of Deungnu (circa 1270) and Gungnu (1354), the water clocks produced during the Yuan Dynasty (1271–1368). These clocks influenced Heumgyeonggaknu (1438) of the Joseon Dynasty (1392–1910), exemplifying the automatic water clocks of the Yuan Dynasty. Deungnu, Gungnu, and Heumgyeonggaknu can be considered as automatic mechanical clocks capable of performances. The Jega-Yeoksang-Jip (Collection of Calendrical and Astronomical Theories of Various Chinese Masters) contains records of Deungnu extracted from the History of the Yuan Dynasty. We interpreted these records and analyzed reproduction models and technical data previously produced in China. The time signal device of Deungnu featured a four-story structure, with the top floor displaying the four divine constellations, the third floor showcasing models of these divinities, the second floor holding 12-h jacks and a 100-Mark ring, and the first floor with four musicians and a 100-Mark Time-Signal Puppet providing a variety of visual attractions. We developed a 3D model of Deungnu, proposing two possible mechanical devices to ensure that the Time-Signal Puppet simultaneously pointed to the 100-Mark graduations in the east, west, south, and north windows: one model reduced the rotation ratio of the 100-Mark ring to 1/4, whereas the other model maintained the rotation ratio using four separate 100-Mark rings. The power system of Deungnu was influenced by Suunuisangdae (the water-driven astronomical clock tower) of the Northern Song Dynasty (960–1127); this method was also applied to Heumgyeonggaknu in the Joseon Dynasty. In conclusion, these automatic water clocks of East Asia from the 13th to 15th centuries symbolized creativity and excellence, representing scientific devices that were the epitome of clock-making technology in their times.